|
Secondary predicates and complex predicates
In Croft’s approach, described in section 3.2, adverbs appear to belong within modifica- tion rather than predication. This conclusion is complicated by the fact that Croft (2003) defines modification as “a secondary propositional act function which can aid to establish reference (restrictive modification) or assert a secondary predication (nonrestrictive mod- ification)” (2003: 184–185). Restrictive modification is clearly instantiated by attributive adjectives. But Croft does not state explicitly how nonrestrictive modification is attested. Since nonrestrictive modification asserts a secondary predication, it could be instantiated
by depictives and resultatives, but adverbs are of course also candidates.
In later work, Croft (in prep.) discusses the matter from a different perspective, where manner adverbs along with resultatives and depictives are placed under the label com- plex predicates. In general, there appears to be little consensus regarding how com- plex predicates are to be delimited (cf. Alsina et al. 1997 and Amberber et al. 2010). Schultze-Berndt & Himmelmann (2004) distinguish depictives from complex predicates, since “depictives constitute a predication which is to some extent independent of that of the main predicate” (2004: 69). In contrast to secondary predicates more generally, it could then be argued that for complex predicates, it is not possible to distinguish between the primary and secondary predicates. Serial verb constructions constitute one typical example of a construction type where the primary and secondary predicates cannot be distinguished. But Croft includes many other construction types in his wide definition of complex predicates: they “are predicates that are expressed by more than one element in a clause” (in prep: 202). Apart from the traditional examples of complex predicate types, such as constructions with serial verbs or deranked verbs, Croft argues that com- plex predicates include various other constructions involving manner adverbs, predicative adjectives, predicative nouns, and so on (in prep: 202ff.).
Within the domain of complex predicates, Croft (in prep.) separates what he terms manner constructions from secondary predicates, which comprise depictives and resul- tatives. While manner adverbs are not explicitly defined as secondary predicates, Croft groups all three construction types together as one “semantically and typologically co- herent subtype of complex predicate” (in prep: 205). The characteristic of this subtype of complex predicates is, according to Croft, that they combine an action concept and a stative concept. This means that manner adverbs, like depictives and resultatives, are “stative predicates” (in prep: 231). But typical manner adverbs such as quickly and slowly can hardly be classified as states. Something cannot be in a state of being slow – rather, moving at a slow pace is dependent on an action concept, denoted by the main event. It appears that properties of events are, or can be, very different from properties of en- tities, which are clearly states (see further discussion in chapters 8 and 9). Croft (in prep.) further discusses the results of Loeb-Diehl (2005) and Verkerk (2009), which show that the encoding of secondary predicates and manner adverbs recurrently overlaps cross- linguistically. Such findings, he argues, support the treatment of these constructions types as belonging to the same type of predicate (in prep: 205). Also Croft points to the neces- sary simultaneity of the property encoded by the manner adverb and the event encoded by the main predicate (as shared with depictives), and emphasizes event orientation as the primary characteristic of manner adverbs. Along with the English encoding difference of secondary predicates and manner adverbs, this is argued to be the main reason why manner adverbs are commonly placed outside the realm of secondary predicates, or even that of complex predicates: “Manner adverbs are often not treated as parts of complex predicates, although as event modifiers, they are no different from object (noun) modifiers in complex (multiword) argument phrases” (in prep: 231). It is interesting, if somewhat problematic, that manner adverbs are treated on the one hand as modifiers on a par with modifiers within referring expressions, and on the other as more or less equivalent to secondary predicates. The question arises as to whether something can be a modifier and a predicate at the same time, and if so, how this is compatible with Croft’s model
of semantic classes and discourse functions (cf. tables 2.2 and 3.1). Moreover, although Croft describes the shared characteristics of depictives, resultatives, and manner adverbs, along with their overlaps in encoding strategies, as sufficient grounds for treating them together as one coherent complex predicate type, he does not explicitly classify manner adverbs as secondary predicates.
In conclusion, opinions differ in the literature on whether or not manner adverbs should be defined as secondary predicates. With support in cross-linguistic findings, Loeb-Diehl (2005) and Verkerk (2009) argue that what they term manner predications are instances of secondary predication. For Himmelmann & Schultze-Berndt (2005b), the difference between participant and event orientation is the distinguishing point between secondary predicates and manner adverbs. Croft (in prep.) partly avoids the issue of whether manner constructions are secondary predicates by focusing on complex predicates, very broadly defined.
Do'stlaringiz bilan baham: |
